England's Top Civil Judge Says AI Should Decide Simple Court Cases

England's most senior civil judge, Sir Geoffrey Vos, says simple civil disputes—personal injury claims and low-value professional negligence cases—could be adjudicated by artificial intelligence, delivering justice that costs far less and resolves far faster than the courts can today.

Vos, the Master of the Rolls, retires from the post in days. As one of his final acts in office, he is pressing the profession to debate whether and how judges should use AI. Current guidance in England confines judicial AI use to administrative tasks. Vos told WIRED he follows that line himself, using Microsoft Copilot daily to locate documents and summarize large bodies of text.

His conviction hardened after a visit to a court in Shenzhen, China, where civil disputes are already decided by judges working with heavy AI assistance. Impressed by what he saw, Vos believes courts across the globe should consider following suit.

"What most people desperately want in civil disputes is resolution," Vos told WIRED. "The question is, where is the line to be drawn?"

The stakes for the justice system are concrete. County court filings in England are rising roughly 10 percent each year, and Vos expects that growth to continue—a surge he attributes directly to chatbots such as Gemini, ChatGPT, and Claude, which any litigant can now ask to draft a particulars of claim or a claim form.

What could AI actually decide?

Vos draws a clear boundary between case types. Complicated commercial disputes and sensitive family or custody cases should stay with human judges, in his view. He sees essential human judgment surviving there even among the younger generation.

But he argues that personal injury claims, low-value professional negligence cases, and similar matters could easily move to AI adjudication. Complex commercial work would not disappear from human hands; instead, AI would collate information to make the decision-making process better and easier.

"We will allow more people access to justice more cheaply and more quickly using technology," Vos said. "That will be an overall good for the economy and for the people we serve."

He frames the shift as a matter of cultural fit rather than technological capability. People already ask large language models the answer to almost every question, including legal ones. A 21st-century justice process that ignores that habit, he argues, will fail.

Consent is his condition. "It will only be possible to ask people to accept algorithmic decision-making if they consent to the process in a particular case, or if society decides through legislation," Vos said. He wants an open, international discussion on how disputes should be resolved.

Why public acceptance is the real barrier

Vos identifies acceptance, not accuracy, as the central obstacle. Human-delivered adverse outcomes, even ones a litigant regards as wrong, can be accepted over time when delivered in a human way. A machine-driven adverse decision may prove harder for people to swallow.

On explainability—the objection that no one can trace how a model reached its verdict—Vos is unmoved. "I don't think it's about reasoning; it's about outcomes," he said. "The AI can produce very plausible reasoning. You can't see how it produced that reasoning, but you can see the reasoning it produced, and you can judge the plausibility."

He does not predict the extinction of the bench. "I don't think you'll ever do away with human judges," Vos said. "There's something almost biblical about the human judge; it's in the human psyche to want to be judged by their fellow man." But he added that in some situations people would rather have a quick outcome, even if it is not a human one.

Vos declined to fix the dividing line himself. "I don't think judges are yet in the position to say. It's a question for society, politicians, and legislation to resolve," he said. "That debate is really important."

How does the top judge use AI himself?

Vos, who spoke to WIRED at his office at the Royal Courts of Justice in London in October, described a working routine built around Copilot. He holds 11 meetings a day. When he knows a relevant document was created years ago, Copilot finds it in seconds.

He does not use AI to write judgments. He remains cautious about using it to read papers for his cases—though he attributes that caution to the fact that he handles the most important cases in the system.

His career spans the profession's entire technological arc. "When I started as a lawyer, we had foolscap paper, typists, and Tipp-Ex," Vos said. "We moved rapidly into word processors, faxes, and eventually computers."

Lawyers, he said, are conservative with a lowercase c—cautious adopters, not resisters. "It means they should be adopting change cautiously, with their eyes open," he said. "The legal system is for the people of this country. You can't be effective in providing access to justice if you're not on the same technological wavelength."

He warned that public confidence rests on judicial integrity, and integrity requires transparency about technology. "In order not to lose that integrity, you have to explain how you're adopting technology and achieve a broad consensus that it is appropriate," Vos said.

Are chatbot-drafted claims flooding the courts?

The 10 percent annual rise in county court cases tracks with the spread of consumer chatbots. Vos does not assume the claims are junk.

"We don't have the data. But I think it's a mistake to assume that just because a claim is created by AI, it's totally frivolous," he said.

He reads the surge partly as good news. Access to justice has historically favored the rich; technology is widening it. "If people have a problem sufficient to bother ChatGPT, they probably genuinely think they have a claim," Vos said. "People are not generally time wasters. They're not generally trying to make trouble."

The claim flood carries real costs. Multiple lawyers have been caught citing case law hallucinated by AI. In a first for the UK, someone recently lost their license to practice in part because they misused the technology. Litigants acting as their own counsel with chatbots are filing an unprecedented number of claims, some of them bogus.

What about deepfakes and fabricated evidence?

Vos dismisses the alarm over AI-generated images, video, and voice recordings as a new label on an old judicial task. Judges decide whether people are telling the truth every day, and humans, as he put it, are not bad at lying.

"There's absolutely no difference—in scale or qualitatively—in somebody producing a false image created by AI," he said. "The only reason we're all getting so excited about deepfakes is because they are so new. But actually, that is the stuff of judging."

Courts have always tested forged wills and doctored evidence, he argued. AI fabrication extends the category, not the function. His warning to would-be abusers was blunt: "People will find out when they produce fake images to the court that they are committing perjury and they will go to prison, just as if they'd forged a will."

"I think we need to be cautious about predicting disaster. I'm very optimistic," Vos said.

What happens after Vos retires?

For Vos himself, the debate is now philosophical; he leaves it, as he said, to a new generation of leadership judges. For everyone else in the profession—and for coders, cybersecurity experts, math professors, and others—the rules are changing fast.

His parting position sets the terms of the argument: the technology can already adjudicate simple disputes, the public has not yet consented to let it, and only legislation or explicit case-by-case agreement can close that gap.